Attribute not being searched

I have an index with an attribute called page_paragraph that has been configured as a searchable attribute. Algolia however does not search at all on this specific attributes. It does return search results that match other attributes in that list.

Hi @ambarish,

Can you give us a bit more information here so that we can help you out?
Maybe an example record + settings configuration could help.

Looking forward for more context :wink:

You have full permission to access my Algolia account - JD256SSXSD (

  1. Index: rcdocsindex. This index has 178 records.

  2. JSON blob -

“page_name”: “deploy/single-pane-of-glass-spog”,
“main_section”: " Deploy",
“sub_section”: " Before you start",
“page_heading1”: “Single Pane of Glass (SPOG) View”,
“page_heading2”: “”,
“page_heading3”: “”,
“page_heading4”: “”,
“page_heading5”: “”,
“page_heading6”: “”,
“page_heading7”: “”,
“page_heading8”: “”,
“page_heading9”: “”,
“page_heading0”: “”,
“page_link1”: “”,
“page_link2”: “”,
“page_link3”: “”,
“page_link4”: “”,
“page_link5”: “”,
“page_link6”: “”,
“page_link7”: “”,
“page_link8”: “”,
“page_link9”: “”,
“page_link0”: “”,
“page_paragraph”: "The Single Pane of Glass (SPOG) view is a real time, interactive, visual representation of all pipelines across your organization. Each job in this view will update colors in real time based on the status of the job: You can interact with the SPOG view in the following ways: ##Zoom The SPOG view is meant to give you one dashboard where you can have a macro view across all pipelines in your organization. However, you might want to focus on specific sections of your pipeline in some cases, like seeing a ‘red’ job in the pipeline for example. You can zoom in or out of any part of the pipeline with your mouse scrollbar or with pinch-to-zoom on your mouse keypad. A zoomed representation of the pipeline is shown below: ##Run/pause job You can right click on any job to trigger a manual run or pause the job. Please note that if you choose to trigger a manual run, the job will run with the latest input values available to it. Pausing a job turns off automatic triggers for the job. You can continue triggering paused jobs manually. ##View console output You can click on any job in your pipeline to view console logs. Please note that in most cases, the logs you are interested in will be under the Executing Task → /home/shippable/micro/nod/stepExec/managed/ / section. This section is shown as expanded in the screenshot above. In addition to viewing logs for the latest run, you can also view logs for historical runs by choosing a past run in the UI. ##Tracing upstream events and data The first question people tend to ask when their application breaks is: what changed? Typically, the more complicated a pipeline gets, the more difficult this question becomes to answer. To make this easy on the user, Shippable has embedded trace functionality in every pipeline job. It allows you to pick any particular job in your pipeline, and view the entire tree of events that led up to that particular version of that particular job with just a few clicks. This allows you to go all the way back to the initial code commit that triggered the workflow to begin with. The trace view shows the inputs to a given job. To get to the trace view: For example, let’s assume your deploy job just ran and you want to see what was deployed. When you click Trace , the list shows the inputs to that deploy job: In the image above, the deploy job has two inputs: a manifest job and a cluster resource . You can click on the + icon on the left of the manifest name to expand it and see which image and tag were deployed. ##View in tabular format The default view of your SPOG is graphical. However, you might want to see all jobs are resources in a table in order to quickly find a particular job or resource. You can switch between table and graphical format by clicking on the Grid View button on your SPOG: You can take some actions on the jobs in grid view, like running or canceling builds, soft or hard deleting resources and jobs, or viewing console logs for jobs. ##Filter SPOG If you have a lot of pipelines configured, the SPOG view can get busy and it can be difficult to find a specific job or resource. You can use the flags keyword to filter SPOG for easier readability. The flags keyword can be set in any job or resource: Once you commit this, the filters will be shown in the Filter dropdown in your SPOG view. You can choose one or more filters as shown below: Please note that you do not need to add flags to every job or resource in your pipeline. If we detect flags for a job or resource and you filter SPOG with that flag, the UI will display all upstream and downstream jobs and resources in the pipeline. ##View orphaned and soft-deleted resources Orphaned resources are resources that are defined in your shippable.resources.yml but not used as inputs or outputs for any job.\nSoft-deleted resources and jobs are resources/jobs that are deleted from your yml files. We do not automatically delete these to avoid situations where something gets mistakenly deleted from the yml and you want to retrieve it at a later time. To view either or both of these types of resources, click on the arrow dropdown in your SPOG pane header and choose the ones you want to view. ##Testing your pipeline config Dry Run feature allows you to validate and test your syncRepo pipelines before actually adding them. Currently we support dry-run pipelines for branches only, dry-run pipelines for Pull Requests is work in progress. To check dry run pipeline for a branch, goto subscriptions pipeline tab. On the right, click\non Dry Run button. Search and select the project for which you want to check dry run pipeline. After selecting the project, you should select a branch for which you whould like to view dry run pipeline. After selecting the branch, your dry run pipeline will show up. It will look exactly same as how it will look if you add an actual syncRepo for selected project and branch.\nAny errors or warnings will be shown to you so that you can correct them before adding syncRepo. To check the dry run pipeline for any other project or branch, you can select the same from dropdowns and dry run pipelines for those will be shown. Please note that, since we support resource sharing between syncRepos in same subscription, if any step(IN, OUT or NOTIFY) resources or jobs for a job are not present in current syncRepo but are present in other syncRepo in the same subscription then those resources or jobs\nwill be shown in dry run pipeline. ",
“page_listitem”: "Green if the last run was successful Red if the last run failed Blue if the job is in progress Grey if the job is queued and waiting for available capacity Zoom in and out to see additional details for any part of the pipeline Run or pause the job View console output and additional meta-data for each version of the job. Trace upstream meta-data View in tabular format Filter with flags View orphaned and soft-deleted resources Click on a job in the SPOG to open the console. Click the Trace tab in the left sidebar menu. This opens a list of the job’s inputs. You can drill down to see the versions of the inputs and all data for that version. ",
“page_code”: "jobs: - name: type: #other job related inputs flags: - ",
“page_baselink”: “”,
“page_map”: {
“page_heading1”: “
“objectID”: “2083330”

  1. page_paragraph is not being returned in the API response nor in the dashboard for any matching term.

Settings -

  1. page_paragraph is in the searchable attributes.
  2. page_paragraph is also specified in the list of attributes to return.

Please let me know if you need any other details.

Also - please let me know how I can make all my attributes in my JSON blob searchable. My JSON blob is dynamically constructed and the attributes/properties are not static. So i want to be able to specify something like make all attributes searchable rather than painfully selecting each and every attribute.

Hi @ambarish,

I wasn’t able to find any rcdocsindex index nor a page_paragraph inside the rctestindex. Did you change something?

By default if you don’t configure any searchable attribute, all records strings attributes are considered as searchable. But then the relevance might not be pertinent.

Do you still have an issue within your rctestindex? You say that your json blob is dynamically constructed, but don’t you have a way at the same time you’re constructing it, to dynamically set the searchableAttributes?

1 Like